TOPLINE:

Bending forward ≥ 30° during pregnancy at an occupation is associated with more than a 35% increased hazard for miscarriage per additional hour, while walking shows an 18% increase and standing a modest 3% increase.

METHODOLOGY:

• Researchers analyzed 803,829 pregnancies in 475,312 employed women in Denmark between 2004 and 2018, retrieved from the DOC*X-Generation (DOC*X-G) national register-based occupational cohort.
• Occupational exposure to standing, walking, and forward bending ≥ 30° was assessed using the PRECISE job exposure matrix (JEM), a quantitative, pregnancy-specific matrix based on accelerometer measurements from 403 female workers and expert ratings, matched to women's occupational codes at pregnancy start.
• Miscarriage was defined as pregnancy loss before gestational week (GW) 22, with information retrieved from the Danish National Patient Registry (DNPR) using International Classification of Diseases (ICD-10) codes DO00-DO07.
• Adjusted hazard ratios (aHRs) with 95% confidence intervals (CIs) were estimated using Cox regression with GW as the underlying timescale, following women from GW 4 until miscarriage or reaching GW 22.
• Models were adjusted for calendar time, maternal age, previous miscarriages, parity, educational level, and country of birth, with being absent from work the preceding week assessed as a time-dependent effect modifier.

TAKEAWAY:

• Each additional hour of occupational standing was associated with an aHR of 1.03 (95% CI, 1.02-1.04) for miscarriage in the fully adjusted model, with the aHR increasing to 1.05 (95% CI, 1.03-1.07) in women who were absent the preceding week.
• Each additional hour of occupational walking was associated with an aHR of 1.18 (95% CI, 1.15-1.22) for miscarriage, with aHRs of 1.15 (95% CI, 1.11-1.19) in women who were not absent and 1.19 (95% CI, 1.12-1.28) in those who were absent.
• Each additional hour of occupational forward bending ≥ 30° was associated with an aHR of 1.36 (95% CI, 1.30-1.41) for miscarriage, with aHRs of 1.29 (95% CI, 1.23-1.35) among women who were not absent and 1.42 (95% CI, 1.30-1.54) in those who were absent the preceding week.
• Post hoc quartile analyses indicated a consistent exposure-response relationship only for forward bending, while associations for standing and walking decreased at the highest quartile, suggesting potential healthy worker selection bias.

IN PRACTICE:

"In Denmark, there are currently no formal guidelines addressing occupational standing or walking within the first 4 months of pregnancy, and no guidance regarding forward bending. Our findings highlight the importance of including the early pregnancy stage into guidelines for pregnant workers," wrote the authors of the study.

SOURCE:

The study was led by Hannah Nortoft Frankel, MD, PhD, Department of Occupational and Environmental Medicine, Bispebjerg Hospital in Copenhagen, Denmark. It was published online in Occupational and Environmental Medicine.

LIMITATIONS:

The lack of individual-level information on smoking during pregnancy among women with registered miscarriage is a major limitation, as smoking is a well-known risk factor. The study may be subject to healthy worker selection bias, as job changes or workforce exit before pregnancy due to health issues may have attenuated results, and annual job code recording did not capture job changes during pregnancy, potentially leading to differential misclassification. Residual confounding may be present from correlated occupational co-exposures, including night-shift work and chemical exposures, and the study was unable to account for potential confounding by other maternal health conditions and behaviors. Detection bias regarding registration of miscarriages may exist, particularly for very early miscarriages, which are more likely to be detected in women with higher socioeconomic position and lower levels of occupational physical activity. The attenuation observed in the highest quartile for standing and walking raises uncertainty about the results at the highest exposure levels and could reflect healthy worker selection bias.

DISCLOSURES:

The research received support from the Working Environment Research Fund (grant number 24-2021-04) and the A.P. Moller Foundation (grant number L-2023-00086). No competing interests were declared by the authors.

This article was created using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication.